IEEE Workshop on Automatic Speech Recognition and Understanding, 2005. 2005
DOI: 10.1109/asru.2005.1566530
|View full text |Cite
|
Sign up to set email alerts
|

Evaluation of natural emotions using self assessment manikins

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
94
0
1

Year Published

2007
2007
2017
2017

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 142 publications
(95 citation statements)
references
References 3 publications
0
94
0
1
Order By: Relevance
“…The "large-size fonts" (27 points and higher) have similar impact with "small size fonts". Table 1 The statistical analysis of our results, is based on the approach used by (Grimm & Kroschel, 2005) on listeners' emotional state assessment. From equations (1), (2), (3) and (4), the resulting mean values of the participants-evaluators confidence scores ( k r ) are shown in Table 2 Fig …”
Section: Resultsmentioning
confidence: 99%
“…The "large-size fonts" (27 points and higher) have similar impact with "small size fonts". Table 1 The statistical analysis of our results, is based on the approach used by (Grimm & Kroschel, 2005) on listeners' emotional state assessment. From equations (1), (2), (3) and (4), the resulting mean values of the participants-evaluators confidence scores ( k r ) are shown in Table 2 Fig …”
Section: Resultsmentioning
confidence: 99%
“…Then he/she was asked to select the best describing image for each primitive. This evaluation method yields one reference value for each primitive i ∈ {valence, activation, dominance} and each utterance n. The individual listener ratings were averaged using confidence scores as described in (Grimm & Kroschel, 2005a). One half of the database was evaluated by 17 listeners, the other by 6 listeners, which was due to the fact that the second half of the database was recorded and evaluated later, when only a smaller number of evaluators was available.…”
Section: Emotion Evaluationmentioning
confidence: 99%
“…A group of evaluators listened to the emotional sentences and assessed the emotional content. For this human evaluation we used the Self Assessment Manikins (Fischer, et al, 2002;Grimm & Kroschel, 2005a). In this method five images were offered per emotion primitive.…”
Section: Emotion Evaluationmentioning
confidence: 99%
“…All annotators evaluated all clips, with five (out of seven) performing the entire process twice to allow the validation of intraannotator agreement. Expected emotion was derived from the individual experienced emotion annotations using a correlation-based scheme similar to that in [78] with particularly uncorrelated annotations being rejected as outliers. Intended emotion annotations were performed three times by a single expert, expected to be very consistent.…”
Section: Emotion Annotationmentioning
confidence: 99%